Skip to content

Guides

Check if an IP, domain or file hash is malicious

You have an indicator - an IP from a firewall log, a domain from an email, a hash from an EDR alert - and one question: is it bad? Here is how to answer it with a free API call, and what the answer does and does not mean.

One call

Send the indicator to the search endpoint (free keys are 100 credits a day, no card - get one here). The response includes an indicator block saying whether the value appears in validated incident reporting, with the feed confidence:

curl -s "https://threatcluster.io/api/public/v1/search?q=45.153.229.109" \
  -H "X-API-Key: tc_live_..."
{
  "indicator": { "value": "45.153.229.109", "seen": true, "confidence": "high", ... },
  "clusters": [ the incidents it appeared in, with sources and dates ]
}

Works the same for domains, URLs, MD5/SHA1/SHA256 hashes. The answer carries its evidence: the clusters tell you which campaigns the indicator was reported in, not just a verdict.

"Not seen" is an answer too

"seen": false means the value has not appeared in validated reporting - it does not prove the indicator is safe. Treat it as "no public evidence", and check a reputation database (AbuseIPDB for IPs, VirusTotal for hashes) for crowd-sourced signal. The two kinds of sources disagree usefully: reputation databases have breadth; incident reporting has context.

Bulk and blocking

Checking one indicator is a lookup; blocking thousands is a feed. The public IOC feeds carry the current high-confidence set as plain text, CSV, JSON, hosts, AdGuard, Suricata and MISP - no key needed. Every indicator in them links back to the reporting it came from.

For AI agents

The same check is an MCP tool (search_everything): npx -y threatcluster-mcp, or the one-click Claude connector. An agent triaging an alert can resolve "is this IP bad" with one tool call and cite the incidents.